« Connecticut legislature sends death penalty repeal bill to Governor Rell | Main | Connecticut Governor pledges to veto death penalty repeal bill »

May 22, 2009

"Is 500 serious crimes worth the freedom of 50,000 offenders? That's my question."

At the end of a lengthy debate in the comments to an earlier post, frequent commentor "federalist" tees up a great question that goes to the heart of the challenges facing all sentencing reform efforts.  The question federalist poses is the question in the title of this post, and I wanted to set it out in this new post so that it could be discussed and debated in a fresh setting. 

My students know that I say that the right answer to every good question is "it depends," and I think that this is the right answer to federalist's great question.  First, of course, we need some sense of what "serious" crimes are hypothetically being prevented by denying freedom 50,000 offenders.  I think most everyone would say that heroin dealing and downloading child porn and illegal dogfighting and mortgage fraud and perjury and domestic violence are all serious crimes, and yet I suspect few would endorse trading off freedom for 50,000 offenders to prevent 500 of these kinds of "serious" crimes.

To his credit, federalist provides a sketch of his vision of  "serious" crimes: "forcible rape, armed robbery, murder, drunk driving."  I suspect he added drunk driving to the list because I often complain that we do not seem to treat drunk driving as a serious crime unless and until the drunk driver kills or seriously hurts someone.  If we accept the idea that drunk driving is a serious crime, I wonder if federalist and others would seek to deny the freedom for Charles Barkley and Joba Chamberlain and Mel Gibson and Lindsay Lohan and Nicole Richie and 49,995 other recent drunk driving offenders in the hope of preventing 500 new cases of drunk driving.

Even leaving drunk driving out of the debate, we also have the fundamental question of whether this hypo sets up a false choice because of the other costs of denying freedom to 50,000 offenders.  Specifically, at current rates, it could cost well over $2,000,000,000 deny this freedom for just one year.  I would think spending that $2 billion on additional cops (or better health care and schooling) could prevent even more serious crimes.  Plus, of course, there is the (abstract?) cost to freedom and liberty for the 49,500 offenders (and all who know them) from being locked up to prevent others from committing crimes.

That all said, while we could seek to further refine or dispute federalist's basic great question, modern risk assessment tools and crime data create the need for all would-be sentencing policy-makers and reformers to confront some variation of this great question at some point in some way. 

So, dear readers, how do you respond to it?  Relatedly, how do you think Thomas Jefferson and the authors of the Federalist Papers and others who founded the US would respond?

May 22, 2009 at 09:52 AM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference "Is 500 serious crimes worth the freedom of 50,000 offenders? That's my question.":


There is, of course, something else in the mix, retribution.

As for drunk driving, I think that house arrest, ignition locks etc. need to be used more. The NM stats you cited are eye-opening. There ought to be graduated penalties.

And you're right, the drunk driving was tongue-in-cheek . . . .

Posted by: federalist | May 22, 2009 10:09:48 AM

After we incarcerate 50K what are you going to do about their replacements. The incapcitation plan sounds gulag-like. At what point are we going to start freeing the 50K when they reach 35, 40, 45 years old --never?

Posted by: yo dog | May 22, 2009 10:13:32 AM

I think Thomas Jefferson would respond very simply.

Innocent until proven guilty.

That's the fundamental flaw in any notion of "future dangerousness." In the future everyone is innocent. Completely, wholly, totally innocent.

Posted by: Daniel | May 22, 2009 10:37:28 AM

Why not just make the sentences longer? Or the length of parole?

Posted by: . | May 22, 2009 10:59:00 AM

Yo dog's "replacement" question contains a premise that the number of slots available for criminals is fixed, and there are people waiting to fill them if only the present perpetrators move out of the way.

That premise is plausible for crimes involving a market, such as drug dealing. It is nonsense for murder, rape, robbery, and burglary, where there are plenty of targets for anyone who chooses to commit the crimes.

Daniel, I believe the question is whether and how long to lock up people who have been proven guilty, not preventive detention based on "Minority Report" type predictions.

Posted by: Kent Scheidegger | May 22, 2009 11:17:58 AM

In the federal system, at least, we are discovering that recidivism could be reduced through a reallocation of resources to community supervision of moderate to high-risk offenders (i.e., those who, on the Risk Prediction Index, score somewhere between 5 and 9). Several districts are seeing positive results and are deeply engaged in sophisticated, cutting edge, evidence-based reentry programs that use resources much more effectively than the typical district. The recidivism rate among federal ofenders, according to the BOP, is 45%.

Research indicates that when resources are focused on low risk offenders, they're actually more likely to recidivate. The system incentives (e.g., workload measurement, etc), therefore, need to change for probation and community supervision agencies to take greater steps with moderate to high risk offenders and increased use of administrative caseloads with low risk offenders. It would probably help also if there was greater effort put into preadjudication programs such as diversion, for nonviolent defendants, and preadjudication problem-solving courts which have been shown to be effective.
That said, releasing prisoners without providing the resources at the community level and then using those resources effectively, will result in significant recidivism and victimization.

According to the Administrative Office of the U.S. Courts, a year of prison costs taxpayers about $25,000 per year while a year of community supervision costs about $3,500 per year not even acccounting for whether the individual is becoming a productive member of society through employment, making child support payments, restitution payments, etc.

Looking at this scenario as someone who works in the criminal justice system, who lives in the inner city, and who pays fairly high taxes, I'd say that further investment in smart community supervision, as opposed to the same or greater use of incarceration, would be well worth the return.

Posted by: Moshe | May 22, 2009 11:29:23 AM

The cost argument is what resonates with me: $25,000 per body per year locked up in prison. And we spend a lot of money locking up guys who haven't physically harmed anyone and aren't an ongoing danger to society.

Take Jeff Skilling. Do we really need to keep him in a cage for the rest of his life? (His long sentence, while technically not life, is practically the functional equivalent, given his age.) What exactly is the purpose of that sentence? Is it retribution? In that case, I'd rather see a punishment that costs us a lot less money.

Posted by: Marc Shepherd | May 22, 2009 11:44:06 AM

Kent. You are correct. But the question is whether X crimes are worth the freedom of Y offenders. But "worth" implies that there is some type of price, some type of trade off. But such a trade-off doesn't exist.

The question as put by federalist is gobbledygook. We could rephrase the question this way, "Are five hundred Martian green cheeses worth the freedom of 50,000 offenders?" and the question would make just as much sense. And I have no idea if it's worth it or not because Martian green cheeses don't exist.

Posted by: Daniel | May 22, 2009 12:04:50 PM

Daniel, there is an inherent trade-off, even if some people aren't sophisticated enough to realize it. Our system makes such trade-offs in all sorts of ways. Severe mandatory sentences keep dangerous people locked up, but they also ensnare other offenders for much longer periods than actually necessary.

Posted by: Marc Shepherd | May 22, 2009 12:21:40 PM

As a practical matter to depopulate a prison or jail you have to reduce the prison admission rate. About half of the admissions are returns and revocations and that is the main reason that recidivism prevention is being emphasized. Assuming that recidivism prevention is in place It also helps reduce the population if the prison staff are diligent in identifying prisoners that are are good risks for parole. If you don't have good recidivism prevention the majority of those paroled will return.

If you plot the number of prisoners versus the number of years since admission the resultant curve can be fit with an exponential decay curve. For Iowa violent offenders the half life (time needed to release half the prisoners) is 58 months and for Iowa drug offenders the half life is 12 months. This means that there is insufficient time for treatment and rehabilitation of drug offenders while they are in prison and if there is to be any rehabilitation and treatment it has to be done after release under community supervision.

The length of confinement of prisoners serving normal sentences for violent and drug trafficking offenses does not appear to be differ significantly from that of those serving mandatory minimum sentences. I believe the parole board policies are more important than the type of sentence.

It used to cost $1 million to incarcerate 40 persons for a year and now it costs that much to incarcerate 33 persons. There are city blocks that have that many residents incarcerated so in some neighborhoods we are spending $1 million per year per city block to provide public safety. I think we could provide as much public safety at a much lower cost if we were willing to change our minds and tactics.

Posted by: John Neff | May 22, 2009 2:17:19 PM

First, I'm pretty sure Jefferson et. al. would think that the people who committed the 500 crimes are responsible for them, not the 49,500 others being held, presumably, prospectively out of some vague, petty bourgeois fear.

But before debating the dubious likelihood of originalist support for federalist's position, there's a false premise behind the question that the state can afford to pay to lock up those 50,000 people indefinitely. I don't think that's true.

You have to put a cost figure on it - just for a ballpark, 50,000 x $30K per year would be $1.5 billion per year, or $15 billion over 10 years in current dollars. So that's the biggest "it depends" of all. In California, for example, where annual prison costs are much higher than that and are a premier budget buster, there's little question economics, not moral abstractions, would dictate how the question gets answered in the real world.

Another false premise is that incarceration is the only possible way to avoid those 500 crimes. How many could be avoided by investing in mental health services? In transitional housing for the homeless? In outpatient drug and alcohol treatment? In both your headline and all too frequently in debates over crime and punishment, we have been presented a false choice.

I don't pretend for a moment the public cares about the humanity of offenders, except perhaps in the abstract or in the context of reducing recidivism. But they do care about high taxes and the fact that prisons are too full, too expensive, don't achieve their stated anti-recidivism goals, and contribute too little to economic growth compared to other types of government spending.

Posted by: Gritsforbreakfast | May 22, 2009 2:30:33 PM


Except that those same founders wouldn't have had a problem with simply executing the 50k for the crimes they already did commit. If the 50k are no longer alive, they are extremely unlikely to commit any additional crimes.

Posted by: Soronel Haetir | May 22, 2009 2:58:03 PM

I don't understand the question. What 50,000 people will be incarcerated to prevent what 500 crimes?

The formula One-Two-Three-Dead solves all problems, including the problem of false convictions. If the Third Conviction is false, you are still executing a bad guy.

The count is for a violent offense, with injury. Any crime causing $6 million in damages, including medical costs, is a capital crime. I am conceding the $6million as the value of human life, even though it is really $2 million.

As to drunk driving, look for damages before meting out punishments. If you stop all cars, 10% of drivers are legally drunk, and driving well.

Posted by: Supremacy Claus | May 22, 2009 3:21:41 PM


The idea is how long do you keep 50k convicts locked up in order to prevent 500 recidivist offenses.

Posted by: Soronel Haetir | May 22, 2009 3:26:32 PM

I am dreadfully disappointed that Supremacy Claus managed to get through an entire post without using the phrase "rent-seeking lawyers".

Posted by: Amendment 21 | May 22, 2009 3:54:57 PM

"Daniel, there is an inherent trade-off, even if some people aren't sophisticated enough to realize it."

Sure, there a trade off when dealing with knowns. But the 500 crimes in federalist's question don't actually exist. You can't have a trade-off between something that is defined and something that is undefined. It's dividing by zero. What you get is another unknown, another undefined. Or as my calculator puts it: error.

Posted by: Daniel | May 22, 2009 5:38:20 PM

SH: Thanks. I see it now.

The answer is 101 minutes. This is lawyer safe math. It is at the 4th grade level. Call it a word math problem. (Why does lawyer math stop at the 4th grade, you might ask. It does, because that is the highest grade level needed to count money.)

Everyone in stir has failed to respond to lesser remedies, and will be a busy criminal. You can count of their committing an average of a felony a week, possibly true in stir, as well. 50,000 felons would commit 50,000 felonies a week. To commit 500 crimes would take a 100th of a week (oh-oh, fractions are in the fifth grade. Relax, just move the decimal place 2 places left). A day contains 24 hours times 60 minutes an hour, making 1440 minutes, times 7 days a week, making 10080 minutes. Divide by 100, and you get about 101 minutes.

Posted by: Supremacy Claus | May 22, 2009 7:25:11 PM

As to foreseeability. It is that of planetary orbits, with no year being an exception. This is 100% scienter in the criminal lover, rent seeking criminal cult enterprise. It is a giant intentional constitutional tort. Because black folks carry a 6 fold higher burden of crime victimization, from all time, to the far off future, racial animus warrants exemplary damages against all guideline makers and the government.

The lawyers leading the reparations movement are traitors to the plaintiff class. They have diverted attention to a hopeless case, since no slave remains alive. But crime victims are all around them, or might even be them. It is the law school indoctrination that made bright people into clueless morons.

Posted by: Supremacy Claus | May 22, 2009 7:31:12 PM

I identified with a dissent by Justice Scalia, in a kansas case, it follows:
"No matter how unsettling the results, we do not have one Constitution for good people and another for bad people. The decision that we write today has to cover not only the obviously
unsympathetic defendants we have before us. What we say to these bad people will also apply to other less bad people, and even a few good people."

On these posts, the harsher comments always like to refer to a John Couey who buried an 8 yr old alive. But, the long harsh sentencing laws, and all the laws which are unfriendly to defendants, also trap people who have done things, not so bad, but were poor, maybe mentally ill and did not know how to negotiate the justice system, our laws ensnare some innocent and some "not so bad" along with the horrible people.

Posted by: DL | May 22, 2009 7:46:46 PM

For an somewhat old, but well done professional survey of inmates, rummage in this searchable and openly available book:


If someone knows of a more recent one or an international one, it would interest the readers here.

Posted by: Supremacy Claus | May 22, 2009 7:47:17 PM

DL: You cannot say the V word. There are 24 million a year. You care nothing for them. They pay your salary, and they are totally betrayed.

Posted by: Supremacy Claus | May 22, 2009 8:47:29 PM

Another argument favoring One-Two-Three-Dead. No education in crime in prison, totally protected by the terrorist lover cult crimnal.


Posted by: Supremacy Claus | May 23, 2009 5:44:44 AM

I can say Victim. There are a lot more victims in this life than any figure you calculated, and I do not know how one would calculate victims. What criteria would one use? We sin against our brother, and we ask to be forgiven as we forgive those who sinned against us.

Two wrongs dont make a right. When our society grabs a human being, who doesn't have a lot of money, accuses him of a crime, say for instance a sex crime, and offers him minimal in the way of defense counsel, and then locks him up for 25 years (Jessica's law)I consider that poor guy a V. As for the accuser, is she a victim? With all the laws that have been passed such as rape shield laws, and other unethical laws, along with the public's fanatical thirst for scapegoats, the "V" is unexamined. It only takes an accusation by an immature 15 year old girl, only an accusation, and a 50 year old man is locked away for life if he gets the 25 year minimum. As for the V? Who knows? Who ever knows? Maybe she was a V and maybe she was a liar. The feminist movement has been largely to blame for this situation and, the politicians who pander to them and made laws which were not in the namae of justice. As for rent seeking lawyers, I agree that most prosecutors are rent seeking lawyers, not striving for any sort of justice, but striving for money. As for any lawyer who would defend a poor sucker as an accused sex offender, I cannot see him as rent seeking, he could seek rent doing thousands of things which were less painful than attempting to defend a man accused of a sex crime.

I can say the V word, I've seen many V's locked away for life.

Posted by: DL | May 23, 2009 11:00:24 AM

DL: You do not have to incriminate yourself. Are you a licensed lawyer? If not, you still have the physical ability to utter the V word. It has not been cut out of your vocal cords by criminal cult indoctrination.

If you are not a lawyer, you are just a normal person concerned about justice. We agree completely. I have no argument with you. You have no trouble with the V word.

I have criticized the lawyer for a sickening rate of false positive, false convictions, after spending $millions on accuracy. I have done so just about as often as I have criticized their false negative rate of 99% of crimes going unanswered. If the rate of false convictions is 20% in the death penalty, what could it be in the plea bargain, over in 5 minutes, with no tribunal? The damage to people, families, and entire neighborhoods from lawyer incompetence is outrageous. The sole remedy I can think of is to end all self-dealt, unjustified, unconstitutional immunities. Let torts end lawyer and judge carelessness. I can discuss those later.

Posted by: Supremacy Claus | May 23, 2009 1:28:10 PM

Can't we stop thinking like lawyers, suck it up, and say "Yes".

Posted by: Matt | May 23, 2009 2:12:08 PM

How could torts end it? I am not a lawyer. I am outraged about the fake justice system we have where police and prosecutors wont do anything about most crimes, but go after whoever they think will make them appear to be doing their jobs.

Posted by: DL | May 24, 2009 10:04:55 AM

Let's rephrase the question. Does the fallacy of hasty generalization make good law? DL makes the point and Jessica's law is proof of hasty generalizations. Everyone who uses Jessica's death as an argument does the same. We can be sure the accusation is a hasty generalization because they are never specific. If they were right, they could randomly pick anyone from the Adam Walsh registry and rightfully claim, publicly, that person is a John Chuey, a baby raper and murderer. They don't do so because the individual would consider that fighting words, hopefully though the courts with a defamation lawsuit.

So another way to pose the question would be this: could someone who is not a baby raper and murderer sue and win if publicly accused of being one? Probably, which is why hasty generalizations are always careful to avoid being specific.

Posted by: George | May 25, 2009 12:51:51 PM

But for lazy, cynical law enforcement agents, win-at-all-cost prosecutors, oppressively coercive plea agreements and guideline-hobbled judges the answer might indeed be yes.

I'm surprised some of the lawyers who post here don't seem to know just how easily and often "good people" (innocent or wrongly accused) end up getting railroaded into prison.

Posted by: John K | May 25, 2009 4:38:11 PM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB